savingSessions: tweet analysis

Author

Ben Anderson (@dataknut)

Published

December 12, 2022

1 Background

UK demand response experiments by NG-ESO and retailers such as @OctopusEnergy

Attempt to do some analysis of #savingSession(s) tweets.

2 Code setup

Part of https://github.com/dataknut/savingSessions

Makes use of https://github.com/dataknut/hashTagR, a DIY wrapper for the rtweet rstats package.

data.table   hashTagR    ggplot2  lubridate      readr     rtweet  tidyverse 
      TRUE       TRUE       TRUE       TRUE       TRUE       TRUE       TRUE 
  tidytext  wordcloud 
      TRUE       TRUE

3 Getting data

Grab the most recent set of tweets that mention #savingSession OR #savingSessions OR #savingsession using the rtweet::search_tweet() function and merge with any we may already have downloaded.

Note that tweets do not seem to be available after ~ 14 days via the API used by rtweet. Best to keep refreshing the data every week…

[1] "Found 54 files matching *.csv in ~/Dropbox/data/twitter/savingSessions/"

That produced a data file of 3229 tweets.

We do NOT store the tweets in the repo for both ethical and practical reasons…

Note also that we may not be collecting the complete dataset of hashtagged tweets due to the intricacies of the twitter API.

4 Analysis

4.1 Tweet time line

Figure 1 shows the timing of tweets by hour.

Figure 2 shows cumulative tweets by hour.

Figure 2: Cumulative number of tweets over time

We see roughly the kind of uptick in tweets for Session 2 that we saw for Session 1…

4.2 Content analysis

Let’s try a word cloud.

Inspiration here: https://towardsdatascience.com/create-a-word-cloud-with-r-bde3e7422e8a

Make a word cloud for all tweets

These are unlikely to render the word ‘savingsession’ as it will be in all tweets due to the twitter search pattern used.

We need to remove common words (to, the, and, a, for, etc). These are called ‘stop words’.

Not especially informative…

5 Sentiment analysis

Inspired by https://www.tidytextmining.com/sentiment.html

Take those cleaned words and sentiment them!

The first word cloud are names that have negative sentiment (according to tidytext::get_sentiments("bing")). Remember the size of the words is relative to the count of other negative words.


negative positive 
     286      236

# A tibble: 2 × 2
  sentiment  freq
  <chr>     <int>
1 negative    951
2 positive   1958

The second wordcloud shows positive sentiments for all tweets. Remember the size of the words is relative to the count of other positive words.